Browse Prior Art Database

Method for jumping locations in video playback through user selection

IP.com Disclosure Number: IPCOM000193602D
Original Publication Date: 2010-Mar-05
Included in the Prior Art Database: 2010-Mar-05
Document File: 4 page(s) / 90K

Publishing Venue

IBM

Abstract

When watching video one of the most important functions needed is the ability to navigate forward and backwards through the video. Two of the main methods of providing this is through fast-forward/fast-rewind and next-chapter/previous-chapter functions. This is fine but fast-forwarding/rewinding relies on the viewer being able to watch the video as it's being played at a fast speed, and determine when the appropriate point has been reached, and then stop the fast-forwarding/rewinding. When the video is being played at fast speeds it is harder for the user to identify when that point has been reached as the picture is not on the screen very long. Another problem is that with human reaction times factored in, by the time the user has spotted the point to stop the fast-forwarding/rewinding, the video has moved on significantly and the user has to reverse the direction back to the intended point. This often leads to viewers fast-forwarding/rewinding at much slower speeds in order to make sure they stop at the desired point as it's both easier see when that desired point has been reached, and can avoid too much video elapsing in the time it has taken them to react. When fast-forwarding/rewinding through a long piece of video, this significantly increase the time spent navigating to the intended place. It is also very prone to human error, as at high speeds the desired point can easily be missed. Being able to navigate to the next and previous chapters in a recording is one method for avoiding these problems, however they rely on the video already being split into chapters prior to viewing. This is less useful when users record their own video from, for example, a television channel as a lot of video recorders do not split their recorded programs up into chapters when recording. Even if they do, users may not want to jump to these pre-defined positions, but to other positions in the video. User-defined bookmarks can be created by some video players so that a user can later jump to these positions on a subsequent viewing, but it requires first that the user watch the video and bookmark it appropriately, and so is not very useful for videos that are normally watched once and then deleted. One of the main use-cases for jumping forward and backwards through video recorded from Television channels is to skip commercial breaks. Users could watch the recording and bookmark it for subsequent viewing in order to correctly identify commercial breaks. However as stated earlier, most users will watch the programme only once before deleting it and so bookmarking it for subsequent viewing is not useful in this scenario. Similarly some video recorders will attempt to guess where the commercial breaks are in a recording and bookmark it automatically so that a user can skip to the next chapter on the first viewing. While helpful if guessed correctly, it is easy for the algorithm to be confused by different commercial break patterns, and also requires that functionality present on the video recorder, something that isn't common across many video recorders such as the Sky+ HD video recorder. It also doesn't allow any interaction by the user in order to determine what to bookmark.

This text was extracted from a PDF file.
At least one non-text object (such as an image or picture) has been suppressed.
This is the abbreviated version, containing approximately 26% of the total text.

Page 1 of 4

Method for jumping locations in video playback through user selection

When watching video one of the most important functions needed is the ability to navigate forward and backwards through the video. Two of the main methods of providing this is through fast-forward/fast-rewind and next-chapter/previous-chapter functions. This is fine but fast-forwarding/rewinding relies on the viewer being able to watch the video as it's being played at a fast speed, and determine when the appropriate point has been reached, and then stop the fast-forwarding/rewinding. When the video is being played at fast speeds it is harder for the user to identify when that point has been reached as the picture is not on the screen very long. Another problem is that with human reaction times factored in, by the time the user has spotted the point to stop the fast-forwarding/rewinding, the video has moved on significantly and the user has to reverse the direction back to the intended point. This often leads to viewers fast-forwarding/rewinding at much slower speeds in order to make sure they stop at the desired point as it's both easier see when that desired point has been reached, and can avoid too much video elapsing in the time it has taken them to react. When fast-forwarding/rewinding through a long piece of video, this significantly increase the time spent navigating to the intended place. It is also very prone to human error, as at high speeds the desired point can easily be missed.

    Being able to navigate to the next and previous chapters in a recording is one method for avoiding these problems, however they rely on the video already being split into chapters prior to viewing. This is less useful when users record their own video from, for example, a television channel as a lot of video recorders do not split their recorded programs up into chapters when recording. Even if they do, users may not want to jump to these pre-defined positions, but to other positions in the video. User-defined bookmarks can be created by some video players so that a user can later jump to these positions on a subsequent viewing, but it requires first that the user watch the video and bookmark it appropriately, and so is not very useful for videos that are normally watched once and then deleted.

    One of the main use-cases for jumping forward and backwards through video recorded from Television channels is to skip commercial breaks. Users could watch the recording and bookmark it for subsequent viewing in order to correctly identify commercial breaks. However as stated earlier, most users will watch the programme only once before deleting it and so bookmarking it for subsequent viewing is not useful in this scenario. Similarly some video recorders will attempt to guess where the commercial breaks are in a recording and bookmark it automatically so that a user can skip to the next chapter on the first viewing. While helpful if guessed correctly, it is easy for the algorithm to be...